ray
01/16/2026, 12:13 AMAI Evy
01/16/2026, 12:13 AMAI Evy
01/16/2026, 12:14 AMresponsiveness value to make the agent wait longer before responding. This value ranges from 0 to 1—lower values mean the agent will wait more and respond slower. [(1) ]()
2. Adjust Interruption Sensitivity
Set interruption_sensitivity to a lower value (or even 0) so the agent is less likely to interrupt. When set to 0, the agent would never be interrupted.
3. Use Prompting
Add specific instructions in your prompt to guide the agent to listen carefully to IVR menus before responding. For example: [(2) ]()
You are going to call to a business and they may have IVR setup.
Listen carefully to their setup and press the right digit to navigate
the IVR to reach the support department.
4. Add a Begin Message Delay
Use begin_message_delay_ms to delay the agent's first message, giving the IVR time to speak first. Valid range is 0–5000 ms. [(1) ]()
These adjustments should help your agent wait for the IVR to finish before taking action.AI Evy
01/16/2026, 12:14 AMAI Evy
01/16/2026, 12:14 AMray
01/16/2026, 12:15 AMAI Evy
01/16/2026, 12:15 AMAI Evy
01/16/2026, 12:15 AMAI Evy
01/16/2026, 12:15 AMbegin_message_delay_ms parameter when creating or updating your agent. [(1) ]() [(2) ]()
When creating an agent:
javascript
import Retell from 'retell-sdk';
const client = new Retell({
apiKey: 'YOUR_RETELL_API_KEY',
}) ;
const agentResponse = await client.agent.create({
response_engine: { llm_id: 'llm_234sdertfsdsfsdf', type: 'retell-llm' },
voice_id: '11labs-Adrian',
}) ;
console.log(agentResponse.agent_id) ;
[(2) ]()
When updating an existing agent:
javascript
import Retell from 'retell-sdk';
const client = new Retell({
apiKey: 'YOUR_RETELL_API_KEY',
}) ;
const agentResponse = await client.agent.update('16b980523634a6dc504898cda492e939', {
agent_name: 'Jarvis',
}) ;
console.log(agentResponse.agent_id) ;
[(1) ]()
Add begin_message_delay_ms to your request body with a value in milliseconds. The valid range is 0–5000 ms. [(1) ]() [(2) ]()
Important note: The value is in milliseconds, so for a 2-second delay, you would set it to 2000. [(3) ]()
This parameter only applies when the agent speaks first. [(1) ]() [(2) ]()AI Evy
01/16/2026, 12:15 AMray
01/16/2026, 12:21 AMAI Evy
01/16/2026, 12:21 AMAI Evy
01/16/2026, 12:21 AMAI Evy
01/16/2026, 12:21 AMAI Evy
01/16/2026, 12:21 AMbegin_message_delay_ms only applies when the agent speaks first. [(1) ]() [(2) ]()
The documentation explicitly states: "Only applicable when agent speaks first."
Since you have the user speak first, this parameter will not have any effect. For your IVR use case where you want the agent to wait and listen before responding, consider adjusting:
- responsiveness — Lower values (0–1 range) make the agent wait longer before responding
- interruption_sensitivity — Lower values make it harder for the agent to be interrupted; setting it to 0 means the agent would never be interrupted
These settings would be more appropriate for controlling how your agent behaves when listening to an IVR system.AI Evy
01/16/2026, 12:25 AMray
01/16/2026, 12:25 AMAI Evy
01/16/2026, 12:25 AMjavascript
import Retell from 'retell-sdk';
const client = new Retell({
apiKey: 'YOUR_RETELL_API_KEY',
}) ;
const voiceResponse = await client.voice.addResource({
provider_voice_id: 'provider_voice_id',
voice_name: 'x',
}) ;
console.log(voiceResponse.provider) ;
Set voice_provider to elevenlabs and provide the provider_voice_id from your ElevenLabs account. For ElevenLabs voices, you'll also need to include the public_user_id (the user id of the voice owner) .
Regarding specifically "v3" or a particular ElevenLabs model version, I don't have explicit documentation on how to specify ElevenLabs model versions (like v3) within Retell. You may want to check with Retell support for details on model version selection.
Check out these potentially useful sources:
- [Add Voice]()
- [Get Voice]()
- [Contact Sales]()
Would you like a general answer based on my knowledge about ElevenLabs API integration?AI Evy
01/16/2026, 12:25 AMAI Evy
01/16/2026, 12:25 AM